Advice/comments on DRO

Then a DRO clearly is waisted money. Except you go to Heidenhain, Sony, Mitutoyo. They will be happy to ripp you off and sell you a temperature-controlled shop, along with new tools.

Nick

Nick, thanks but I think that is what I actually said in my last post

" So for that, forget the DRO and keep up the tried and tested techniques mentioned by Tim, John and Nick."

As my background is in Aerospace and a "few microns" were often considered "a mile" I had forgotten just how much money we spent on obtaining such accuracy. In the end it had little to do with Heidenhain, Sony, Mitutoyo and their but a lot to do with the skill of the individual producing the part and yes we were ripped off consistently. I had also forgotten just how accurate a fit can be achieved with a few basic tools an understanding of the geometry and some care and attention. At least now I know what I really want a DRO for I might stand a chance of picking the right one.

Keith

I believe EMC does closed loop control. The scales have to be fine enough resolution and fast enough signal for the process to work. Talk to Jon Elson over at yahoo groups DIYCNC, he sells EMC support hardware, and runs what he sells. The nut of it is that the answer to OP should have been "not normally", rather than "No".

Adam Smith Midland, Ontario, Canada

"Problem is backlash" is completely correct. EMC will do closed loop, but it won't make a silk purse out of a sow's ear. Ballscrews, and tuning the backlash out of the system is part of the downstroke for closed loop.

Adam

I wanted to repeat that, not to correct or contradict you.

Nick

Sorry Nick, miss-understood your meaning. Having a bad day, got loads of things to do today and woke up with Gout so can't do any of them until the tablets begin to work.

Keith

I think we're in danger of not seeing the wood for the trees here.

A DRO is *all* those things you mention, it's the sum of the parts that's important. It *is* a means of easy accuracy - can you really compare the statistical reliability of having to make X and a half turns of a thimble-sized handle compared with whipping the saddle along until it shows ten thou short of the desired position? Similarly, it's widespread use in a million machine shops is tribute to its convenient use.

As for the accuracy of BW's system, their display unit reads out to half a thou, a feature which I use extensively. As I've said elsewhere recently, with this sort of unit, and in the context of the home workshop (mine, at least!) you don't set the DRO to the nearest half thou and then carve off fifty thou swathes until you get there and then expect to have achieved a required diameter to that degree of accuracy. Machine rigidity, tool flex and half a dozen other factors would soon bugger that plan. It's simply an *aid* to working down to a 'tenth' if that's what you need. You can still expect to apply the same care and skill with sharp tools, an angled top-slide and a quick polish of your specs to squeeze that last few microns!

I see my DRO as a major aid to improving machine tool performance...and I'd never want to go back to being without one. Perhaps it's like other things in life, it's not what you've got, it's how you use it!

--

Chris Edwards (in deepest Dorset) "....there *must* be an easier way!"

The controllers from Eaziform,

have closed loop option, at DIY price.

Stu G.

Are you talking about the M641/M642 or MPC5? Not having it read precisely, I couldn't find a parameter for tolerance. But that is a must. In the PDFs, he's talking about *motor* encoders and does spend only a few words on linear encoders. I doubt that it works unless you hit my nose to the text I missed. The whole problem can be demonstrated with climb-milling. The mill bit will move your table back and forth, the controller has to compensate that

*quickly* and allow some positional (configurable) error, or he will oscillate. This can only be done in the controller, because there is very little time. On a 20MIPS-CPU I can't believe that with 5 axes.

All that requires a highly dynamic setup. Look at servo-motors how fast they accelerate. A stepper is a kindergarden compared to that.

Not wanting to tear the eaziform-boards down! But they are a (DIY-)stepper-CNC.

Nick

No need for an excuse. Neither on a good nor a bad day. It was simply a missunderstanding.

Nick

Although I've got the Newall on my mill, I started off with a 3-axis BW system on my lathe; the 3rd axis was for use on the vertical slide. I was, and still am, very pleased with it, and certainly wouldn't want to be without it. As others have said, DRO on a lathe doesn't perform quite the same as on a mill; the inherent variable flexing of the tool and toolpost with different depth of cut means that you can't just use the DRO for the final sizing - the mic is still required, but the DRO is superb for the rough cuts and is OK for final cuts when set up properly. I've found that the DRO can be relied on for small production runs as long as I take the same sized cuts for each unit. My lathe is relatively large for a model engineer's lathe (6" CH and heavily built Chinese) so the problem might be exacerbated for a lighter one.

I was initially going to xfer between the mill (when I eventually got one) and the lathe, but when I got the mill decided to install a permanent facility. I haven't regretted it.

Norman - I have sent you an eMail direct this morning with a reply flag on, but not getting anything back - did you get it ?

In article , Chris Edwards writes

..... I think you only said that in your off-list e-mail to me (just mention it in case anyone is puzzled).

Actually I have just realised that for curve profiling it might be the repeatability which is more important to me than absolute accuracy. IOW, if I go in by 10.02 mm, I need to know it is exactly 0.02 more than the previous cut at 10.00.

David

Gary

Thanks for that, between yourself and Chris I think you have answered most of my concerns. Always worth having the opinions of those actually using the tool. With the BW system I wasn't doubting their ability to work to +/- .002" but felt in practise they would do much better than that given a fair chance.

I envisage using the system to rough out to +.010" on diameter, measure, re-zero (if tool change etc makes it necessary) and apply the finishing cut using the DRO and expecting to get within .001" of the required diameter. For 95% of what I do +/- .001" will do fine, any issues with this expectation?

I have a set of Chinese scales on my small mill with no integration and found even that a major improvement over the normal handwheel scales. Although I haven't actually cut with it yet, the Anilam system on the larger mill seems to perform very well when checked against some long test pieces of known accuracy. The only issue I have with the Chinese scales is protecting them against coolant ingress. For that reason I would like to go the glass scale route as I can't afford a Newell system; but the room available on the Myford is fairly restrictive and that's where the BW system might score. Although prices do vary they are much closer at this lower end than they were a few years ago and I certainly don't want to "spoil the ship" for a =A3100 or so.

Thanks for your help

Keith

-

Treasurer - Firebird Foundation Inc. - Sorry, I got plenty of nothing! However, got an invite to the Chinese New Year. Boss on this side is Prof of IT

(snip)

no

--

Chris Edwards (in deepest Dorset) "....there *must* be an easier way!"

He can't read semaphore.

.

-- Regards,

John Stevenson Nottingham, England.

Visit the new Model Engineering adverts page at:-

That's how I use it on the lathe - DRO to get close, measure with mic or callipers, set the DRO to the mic reading and take another small cut and measure again, then DRO for the final cut. The final cuts with the DRO are to ensure that the m/c flexing is equal. The glass scales on the Machine-DRO seem very well protected against swarf and lubricant, and appear to be easier to fit than the Newall, but I'll reserve final judgement on that when the job is complete.

The BW system fits very neatly on my lathe and another friend has installed another on his Myford using the same technique that I used. I could send photos of the lathe setup if it would help.

Thanks Gary, that is helpful. Sounds like your final position is the one I was tending towards (BW on lathe, Machine-DRO on the mill).

David